Safe prescribing, administration and monitoring — branching viva

Opening (candidate)

I would treat this as two linked problems: a supratherapeutic gentamicin level that needs immediate action, and a system question about how we prevent high-alert-medicine errors. For the level, I would stop or extend the gentamicin, assess renal function and the child for toxicity, and review the prescribing and monitoring chain. For prevention, I would describe the layered system — decision-supported prescribing, unit-based pharmacists, standard concentrations, independent checks, reconciliation, and monitoring with level-to-action protocols. [9] [6]

Branch A — The level and the action

Examiner: The gentamicin trough is 2.4 mg per litre. What does that mean and what do you do? [9]

Candidate: That trough is well above the extended-interval target of less than 1 mg per litre, so the drug is accumulating and the child is at risk of nephro- and ototoxicity. I would withhold or extend the next dose, check the renal function and review for any clinical sign of toxicity, repeat the level as guided, and then look upstream — was the dose correct for the weight, was the interval right, and was clearance impaired by critical illness, in which gentamicin pharmacokinetics vary widely. I would document the action taken so the level is tied to a decision, not merely filed. [9]

Branch B — Why children, and where errors cluster

Examiner: Why are medication errors more common in children than adults? [1]

Candidate: Because almost every paediatric dose is calculated, not memorised — by weight or body-surface area — and the evidence base is thinner and more often off-label. Kaushal's cohort found medication errors in roughly 6 in 100 paediatric inpatient orders, highest in PICU and NICU, where weights are smallest and infusions most common. Developmental pharmacokinetics add to it: water content, albumin binding, hepatic maturation and renal clearance all change with age, so a dose correct one week may be toxic the next, which is why monitoring carries so much weight in paediatrics. [1]

Branch C — Preventing tenfold errors

Examiner: How does your unit prevent tenfold errors in high-alert medicines? [6]

Candidate: Tenfold errors come from decimal points, trailing zeros, and unit confusion, and they cluster in weight-based and infusion calculations and in high-alert drugs. The defences are decimal discipline — never a trailing zero, always a leading zero — showing the per-kilogram calculation in the order, using standard concentrations and ready-to-administer products to remove compounding error, and applying an independent double-check that re-calculates from scratch for insulin, opioids, concentrated electrolytes, anticoagulants, and chemotherapy. Doherty's five-year review showed these errors reach the patient in a substantial proportion and cause harm in some, which is why the checks are mandatory rather than optional. [6]

Branch D — The layered system

Examiner: If you could implement one change to reduce medication errors in your unit, what would it be? [2]

Candidate: I would resist choosing one, because the evidence supports a layered system. Fortescue's analysis ranked computerised order entry with decision support and unit-based pharmacists highest for prevention, and the Cochrane review of hospital interventions found bundled, system-level changes reduce error rates, while reducing actual harm remains harder to prove. If pressed, I would start with decision-supported computerised prescribing plus a unit-based pharmacist, because prescribing generates the most errors and the pharmacist intercepts them at the bedside — but I would state clearly that single interventions leave gaps that only the bundle closes. [2]

Branch E — Vancomycin and the monitoring principle

Examiner: How does vancomycin monitoring illustrate the principle that monitoring closes the loop? [11]

Candidate: Vancomycin is a narrow-therapeutic-index drug whose efficacy and nephrotoxicity depend on exposure, so it cannot be judged clinically alone. The 2020 consensus guideline shifted monitoring from trough toward the area under the curve over 24 hours relative to the minimum inhibitory concentration, targeting an AUC to MIC ratio of 400 to 600, with the upper bound limiting nephrotoxicity; where trough-only monitoring remains, a trough of 15 to 20 mg per litre approximates the target for serious infection. The level is only useful if it triggers an action — a dose or interval change — so the protocol ties every result to a decision, which is the definition of closing the loop. [11]

Close

Confirm understanding with the team, document the level and the action, report the event through the incident system with root-cause review, and feed the learning back as a protocol change so the same weakness cannot produce the same harm twice. The family receives open disclosure and a clear plan for re-checking the level and the renal function. [6] [11]